UniqAI/nvidia-container-toolkit
1
0
Fork 0
mirror of https://github.com/NVIDIA/nvidia-container-toolkit synced 2024-11-25 21:39:10 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
511 Commits 21 Branches 73 Tags 29 MiB
084eae6e0d
Commit Graph

81 Commits

Author SHA1 Message Date
Evan Lezar
084eae6e0d Fix bug in tegra detection 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-20 14:39:36 +02:00
Evan Lezar
55c1d7c256 Fix assertCharDevice matching on all files 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-20 10:08:00 +02:00
Evan Lezar
c77e86137e Add version output to CLIs 
This change adds version output to the nvidia-continer-runtime,
nvidia-container-toolkit, and nvidia-ctk CLIs. The same version
is used in all cases and includes a version string and a git
revision if set.

The construction of the version string mirrors what is done in runc.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-13 07:31:11 +02:00
Evan Lezar
ff86ecb2a5 Include HasNVML check in ResolveAutoMode 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:55:58 +02:00
Evan Lezar
ad9ec1efae Add HasNVML function to check if NVML is supported 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:55:13 +02:00
Evan Lezar
9db5f9c9e8 Remove unneeded legacy discovery 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:53:52 +02:00
Evan Lezar
e591f3f26b Replace experimental and discover-mode 
These changes replace the nvidia-container-runtime config options
experimental and discover-mode with a single mode config option.

Note that mode is now a string with a default value of "auto"
and a mode value of "legacy" is equivalent to experimental == false.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:53:50 +02:00
Evan Lezar
e0ad82e467 Move ResolveAutoMode to info package 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:28:56 +02:00
Evan Lezar
3a1404f2f4 Move isTegraSystem to internal info package 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:28:56 +02:00
Evan Lezar
cf7bb91481 Update nvidia-container-runtime config options 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:28:56 +02:00
Evan Lezar
ba0e606df2 Use toml unmarshal to read runtime config 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-12 10:28:56 +02:00
Evan Lezar
ae57a2fc93 Merge branch 'CNT-2875/create-specific-symlinks' into 'main' 
Create specific symlinks for CSV mode

See merge request nvidia/container-toolkit/container-toolkit!150
 2022-05-12 05:27:43 +00:00
Evan Lezar
1eb0e3c8b3 Merge branch 'fix-executable-locator' into 'main' 
Fix location of executables in PATH

See merge request nvidia/container-toolkit/container-toolkit!148
 2022-05-12 05:26:22 +00:00
Evan Lezar
675fbace01 Add hook to create specific links 
This change updates the create-symlinks hook to also create symlinks for
libcuda.so, libGLX_indirect.so.0, and libnvidia-opticalflow.so

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-11 16:36:49 +02:00
Evan Lezar
d2516cb5d5 Merge branch 'fix-container-root' into 'main' 
Fix bug in update-ldcache hook when OCI spec contains a relative root

See merge request nvidia/container-toolkit/container-toolkit!147
 2022-05-10 22:01:14 +00:00
Evan Lezar
4696d7ee69 Merge branch 'fix-hook-flags' into 'main' 
Use singular instead of plural for hook arguments

See merge request nvidia/container-toolkit/container-toolkit!146
 2022-05-10 22:00:51 +00:00
Evan Lezar
ef6f48e9f7 Use singular instead of plural for hook arguments 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-10 19:55:31 +02:00
Evan Lezar
088db09180 Use executable locator to find low-level runtime 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-10 15:21:48 +02:00
Evan Lezar
1d2e1bd403 Add lookup.GetPath and lookup.GetPaths functions 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-10 14:52:47 +02:00
Evan Lezar
395f6cecb2 Add GetContainerRoot to oci.State type 
This change adds a GetContainerRoot to the oci.State type to
encapsulate the logic around determining the container root.
This Fixes a bug where relative roots (e.g. as generated by contianerd)
are not supported.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-10 11:48:43 +02:00
Evan Lezar
7574a0d7de Make output of bundle directory a debug message 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-09 09:38:16 +02:00
Evan Lezar
335de5a352 Switch to debug logging when locating runtimes 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-09 09:38:16 +02:00
Evan Lezar
c76946cbcc Add nvidia-container-runtime.runtimes config option 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-09 09:38:12 +02:00
Evan Lezar
785f120c31 Fix form -> from in comment 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-06 13:22:34 +02:00
Evan Lezar
9e46d41dbe Add debug logging when checking requirements 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-05 14:14:01 +02:00
Evan Lezar
9f50ac95c4 Add CUDA ComputeCapability function 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-05 14:09:28 +02:00
Evan Lezar
583793b7ae Add processing for requirements and constraints 
This change adds a Requirements abstraction that can be used to check
an images' NVIDIA_REQUIRE_* envvars against the host properties such
as CUDA version or architecture.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-05 13:43:13 +02:00
Evan Lezar
5d7b3a4a96 Return raw spec from Spec.Load 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-05 13:43:13 +02:00
Evan Lezar
a672713dba Add basic CUDA wrapper 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-05 13:43:13 +02:00
Evan Lezar
8f0e1906c2 Add CUDA image abstraction 
This change adds a CUDA image abstraction that encapsulates
the queries performed on a container image (e.g. envvars) to
check certain CUDA properties.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-05-05 13:43:13 +02:00
Evan Lezar
c224832a6d Add log-level config option for nvidia-container-runtime 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 13:56:17 +02:00
Evan Lezar
5211960fc3 Merge branch 'detect-gpus-flag' into 'master' 
Detect use of --gpus flag in experimental mode

See merge request nvidia/container-toolkit/container-toolkit!125
 2022-04-08 11:18:11 +00:00
Evan Lezar
cfca18a5f8 Merge branch 'refactor-csv-mount-spec-discovery' into 'master' 
Refactor CSV discovery to make char device discovery clearer

See merge request nvidia/container-toolkit/container-toolkit!129
 2022-04-08 10:54:06 +00:00
Evan Lezar
dab6f4b768 Specify --force flag when invoking nvidia-container-runtime-hook 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 12:03:22 +02:00
Evan Lezar
2563c1b87c Export GetDefaultRuntimeConfig 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 12:03:22 +02:00
Evan Lezar
62f608a3fe Make order of discoverers deterministic 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 11:59:26 +02:00
Evan Lezar
2c1e356370 Refactor CSV discovery to make char device discovery clearer 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 11:47:47 +02:00
Evan Lezar
196d5c5461 Move NVIDIA Container Runtime Hook executable name to shared constant 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 11:29:27 +02:00
Evan Lezar
f07d110e85 Use DefaultExecutableDir to determine default paths 
This change adds a DefaultExecutableDir = /usr/bin constant that is used
to construct default paths for executables instead of specifying these
explicitly.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 11:28:03 +02:00
Evan Lezar
1ebd48dea6 Merge branch 'add-symlink-hook' into 'master' 
Add hook create-symlinks subcommand to create symlinks in container

See merge request nvidia/container-toolkit/container-toolkit!121
 2022-04-08 09:14:07 +00:00
Evan Lezar
f7c74d35cc Merge branch 'add-hooks-cli' into 'master' 
Add nvidia-ctk CLI with hook command and update-ldcache subcommand to update LD cache

See merge request nvidia/container-toolkit/container-toolkit!115
 2022-04-08 09:13:39 +00:00
Evan Lezar
0de7491ce3 Merge branch 'check-for-nil-modifier' into 'master' 
Return unmodified runtime if specModifier is nil

See merge request nvidia/container-toolkit/container-toolkit!127
 2022-04-08 09:05:24 +00:00
Evan Lezar
d1a38f10a5 Refactor CSV file parsing 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 08:11:10 +02:00
Evan Lezar
d8109dc49b Add missing close when reading CSV file 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 08:00:37 +02:00
Evan Lezar
67602b28f9 Return unmodified runtime if specModifier is nil 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-08 07:50:40 +02:00
Evan Lezar
907736b053 Inject symlinks hook for creating symlinks in a container 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-07 20:25:55 +02:00
Evan Lezar
95797a8252 Move reading of container state for internal/oci package 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-07 20:25:55 +02:00
Evan Lezar
9c3c8e038a Add cache for mounts 
This change adds a cache to the mounts type. This means that if called to get
a list of folders, for example, the result is reused instead of recalculated.
This also avoids duplicate logging.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-07 20:25:19 +02:00
Evan Lezar
d970d0a627 Add discovery for ldconfig hook that updates the LDCache 
This change adds a discovered hook for updating the ldcache as a container-create
hook. The mounts from a discoverer are inspected to determine the folders that must
be added to the cache using the nvidia-ctk hook update-ldcache command.

This is added to the "csv" discovery mode for the experimental runtime.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-07 20:25:19 +02:00
Evan Lezar
740bd3fb9d Add nvidia-ctk config section 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2022-04-07 20:25:18 +02:00